Translate

Saturday, April 28, 2012

High Intensity Interval Training

Some time ago,  I wrote an article called "Do This and You Will Win".  In that post, I suggested  performing interval training as follows:  "ride absolutely as hard as you can for one minute, then soft pedal as easy as you can for three minutes, repeat 5-8 times or until you think you see Jesus." Assuming that the rider also raced on the weekends, these intervals would be done once a week. 
Since writing that article, I have read numerous critiques of interval training, and heard talk from fellow racers about what they think is the best interval training method.  This made me wonder what is the best interval training protocol or if there even is a best method.  As I think back, I can not even remember exactly the source of my High-Intensity Interval Training (HIT) method.  It certainly didn't come from Lemond's "Complete Book of Cycling."  Nor did it come from Edward 'Eddie B' Borysewicz's "Bicycle Road Racing:  The Complete Program for Training and Competition."  And it definitely did not come from Joe Friel's "Cyclist's Training Bible" (though all are recommended books).  My best guess is that I adopted my interval protocol from an article that I read in Bicycling Magazine, probably a couple decades ago, and it probably was written by Edmund Burke (who was well known for writing science-based training articles).

So my current question is two-fold.  First, is my previously recommended method  for improving cycling performance good advice?  And second, are there other HIT protocols that are equal or better? Or to put it another way, is there an ultimate interval?

The importance of intervals

Before I go straight to the answer, I should first explain the importance of HIT, and why it has value.  It is well established among exercise physiologists and others, that any amount of exercise followed by recovery will increase fitness for sedentary (unfit) individuals.  For beginner and moderately-trained cyclists, increased duration and increased frequency of riding combined with recovery is all that is required to further increase fitness.  Unfortunately, this method of increasing fitness has a ceiling and once this threshold is reached, no amount of increased typical or ordinary riding (aka sub-maximal exercise or below threshold markers) will continue to improve a rider's fitness.  As Laursen & Jenkins state, "in the highly trained athlete, an additional increase in sub-maximal exercise training (i.e. volume) does not appear to further enhance either endurance performance or associated variables such as maximal oxygen uptake (VO2max), anaerobic threshold, economy of motion and oxidative muscle enzymes".   In citing Ben Londeree's research, the authors note that "it appears that once an individual has reached a VO2max >60ml/kg/min, endurance performance is not improved by a further increase in submaximal training volume."  This is not meant to downplay the importance of high-volume training, but to highlight that there is a fixed limit for improving fitness by this method alone.   

 Interestingly, not only is adding a HIT program to your training necessary to reach your full cycling potential, but it can also dramatically speed up the rate at which fitness is achieved.  It is generally believed that it can take several years to go from being an unfit to a highly fit bike racer; however research suggests otherwise.  For example:  Hickson showed that after just 10 weeks of HIT, VO2max could be rapidly increased (+44%; p<0.05) and four of his subjects approached or exceeded 60ml/kg/min in this short time frame.  This does not mean that any cyclist can become an elite rider (>70ml/kg/min for men and >60 for women) in a very short period, though many can become "highly fit" (>60 ml/kg/min for men, >52 ml/kg/min for women) in a relatively short period of time.    Fitness gains initially occur rapidly and  depend on the volume, intensity and frequency of training; as greater fitness is achieved, it appears that the development of the physiological capacities witnessed in elite athletes does not continue to come about quickly.  It may take years of high training loads before an individual reaches his or her full athletic potential through vascular and muscular adaptations.  Outside of developing one's physiology, it can also take years to fully master the sport by developing racing and psychological skills (tactics, equipment, diet, technique, psychology, etc).   

Another interesting and valuable fact about HIT is that, to a significant degree, when compared to aerobic (submaximal) training alone,  it can reduce the exercise time required to achieve or maintain a particular level of fitness, as illustrated by Johnathan P. Little, and especially by Iaia FM, Hellsten Y, Nielsen JJ et al. In short, if you have limited time to exercise/train, then interval training is even more important and efficient for the gaining and maintaining fitness.

I have mentioned  VO2max quite a few times in this article, so it is important to understand a few basics points about it.  VO2max (also maximal oxygen consumption, maximal oxygen uptake, peak oxygen uptake or maximal aerobic capacity) reflects the physical fitness of the individual; it is the maximum capacity of an individual's body to transport and use oxygen during incremental exercise.  While there certainly are other measures of fitness, maximal oxygen uptake (VO2max) is widely accepted as the single best measure of cardiovascular fitness and maximal aerobic power.  In this article, 60 ml/kg/min is used as the critical minimal threshold number for defining a "highly trained cyclist".  For perspective and to see the full spectrum of comparative VO2 max measurements, classified from unfit (32 ml/kg/min) to world class cyclists (90 ml/kg/min) see my article "Comparative Measurements of Maximal Outputs for Cyclist".

There is a reason why I emphasize the distinction between "highly fit cyclists" (VO2max >60ml/kg/min) and lesser-fit cyclists (including sedentary individuals): research shows that "highly fit cyclists" do not respond to exercise stimuli in the same manner as unfit and moderately fit cyclists.  Highly fit cyclists typically will not improve from further increased training volume (with enough volume they will actually get worse), whereas lower fit cyclists will almost always improve with increased training volume.  Also, unfit and poorly fit cyclists can dramatically improve their fitness using nearly any HIT protocol, whereas highly fit cyclists not only exhibit considerably smaller gains, but in many cases they will not make any improvements at all with a HIT protocol that is not correctly designed for them (intensity level,  repetitions, rest duration, and frequency).

Research findings from HIT studies

Below (Table One) are some findings from high intensity interval training studies in sedentary and recreationally-active individuals.  (Source: Laursen & Jenkin's, "The Scientific Basis for High-Intensity Interval Training:  Optimizing Training Programmes and Maximizing Performance in Highly Trained Endurance Athletes."  )  The findings are organized by year of publication, and each study is referenced with links at the bottom of this article.  Collectively, they show that lower fit riders respond well to a wide range of interval protocols - and importantly, many of these studies are the foundation for later research that looks for the best HIT protocols for highly fit cyclists.   The general trend seen in these HIT protocols is to lower the work duration as the intensity is increased.  The studies also increase the number of repetitions as the intensity level is lowered to a specific work duration.  Rest between intervals tends to increase in proportion to the amount of work that is done as the intensity increases, and the total number of repetitions is typically determined by fatigue. Tabata's design is a bit of an exception to this trend with a 20 second work duration and a 10 second rest, but his protocol could be described as one 4 minute intermittent high intensity interval.   


key to the abbreviations used in the charts
Surprises

One of the biggest surprise findings for early researchers (and for myself while researching this topic) was that short 20-30 second HIT could improve both VO2max and 40k time trial results.  By training with short, intense intervals which use primary anaerobic energy (see figure 1 below) a person can achieve significant improvements in long sustained efforts, which largely rely on aerobic energy such as 40 kilometer time trialing.  This isn't exactly intuitive.

 The Ultimate Interval

So, what is the ultimate interval?  First we must ask what is it we are trying to enhance?  As we can see from Figure 1 above, short intense events such as track racing require very high anaerobic capacities and endurance events such as 40k time trialing require very high aerobic capacities.  However, for the purposes of this article, I avoid the topic of "sprinting" (5-15 maximal bursts) and focus on HIT for longer time frames that are commonly used in time trialing and criterium racing.  (Sprint training could be discussed in a future article).   With this in mind, I present a collection of findings from high-intensity interval training in highly trained cyclists, below in Table 2.  These are currently the best and most cited studies that I could find on the topic.  It is also important to reiterate that these studies are on highly trained cyclists; as mentioned previously, they respond to exercise stimuli differently than unfit and recreational cyclists.

So, which is the best?  Let me quote one of the lead researchers, Paul Laursen,:  "It is not possible to unequivocally state that one HIT group improved to a greater extent than the other HIT groups."  ("Interval training program optimization in highly trained endurance cyclists").  However, he does go on to pick out the HIT protocol performed at the intensity of Pmax and a duration of Tmax with a 1:2 work-recovery ratio, as being superior by a small margin to the others.    Ian Dille published an excellent article in Bicycling Magazine titled "The Ultimate Interval", in which he describes this particular HIT protocol in understandable terms.   But to call this particular interval the ultimate interval is a little bit premature.  It is extremely effective, but when we look at the last HIT study in table 2, in only 2 weeks, a group of higher fit cyclists (higher starting VO2max) improved nearly as much as the supposed ultimate interval group did in 4 weeks, while using dramatically different protocols. And CRITICALLY: the importance of sample size calculation cannot be overemphasized. The studies that I have cited all have extremely small sample sizes ranging from only 5 to 23 subjects, with most studies having fewer subjects than fingers on my hands.  Size matters, and confidence in research findings goes up considerably with increased sample size, and down with smaller sample size because of variability between subjects (some test subjects may respond VERY different to protocols than others). Therefore, one should be very cautious to pick out one of the studies as definitely the best or ultimate over the others in my post.    

So, what we see is that there are likely MANY different HIT protocol designs that are equally effective.  The general rule for an ideal HIT protocol appears to be to lower the work duration as the intensity is increased, and to increase the number of repetitions as the intensity level is lowered to a specific work duration.  Rest duration between intervals tends to increase in proportion to the amount of work that is done as intensity increases, and the total number of repetitions is typically determined by fatigue.

HIT applied in the real world

I would generally advise executing a HIT regimen based on research and not on intuition or hearsay.  Any of those shown in table 2 are satisfactory.  However, it can be difficult if not impossible to follow these protocols if a person does not have power meters and access to lab equipment for calculating precise PPO, Pmax and Tmax.  Even if you have a power meter, it can be very hard.  For example to determine PPO, you need to find your highest 30-second power output completed during an incremental test where resistance is increased by 15 watts every 30 seconds, starting at a workload of 100watts.  Your Pmax is calculated by finding the corresponding power output that is measured at VO2max during a progressive exercise test, and Tmax is time to exhaustion at Pmax.  Laursen deemed test subjects fully exhausted when they could not keep their cadence above 60 rpm.  Sounds simple?  No, it's not.


You may be able to estimate a PPO by testing yourself on a stationary trainer and using a powermeter device.  After lightly warming up begin riding at 100 watts. Increase power by 30 watts every minute (you will have to control your wattage by cadence primarily) until you reach exhaustion (failure to maintain a 60 rpm cadence).  Wattage at PPO is going to be higher than at Pmax by a small unknown amount.  Pmax can only be determined accurately in a lab environment that can measure VO2max with it's corresponding watt output.   However, you can estimate you Pmax by taking the average wattage produced over a 5 minute maximal effort and multiply that by .934 (based on writings from Andrew Coggan and only applies to highly fit cyclists).

Without a power meter you could just simply follow my simple "no tool" method (other than a watch).  I can not unequivocally state that this protocol is as good as the proven studies below, but based on the principles of HIT it should produce comparably similar results.  All you have to do is ride as hard as you can for one minute and rest three minutes or until you subjectively feel recovered and do it again and again until exhaustion or you see Jesus.  When you see Jesus, that's when you know it's time to stop.

Training volume considerations with HIT

All of the highly trained cyclists in the studies cited maintained a high-volume low intensity (submaximal) training ranging from 285 kilometer plus minus 95 kilometers (177 miles plus minus 59 miles) per week, both before and during the study.  In a review of HIT research Laursen (see 23 below) states that, "a polarized approach to training, whereby about 75% of total training volume is performed at low intensities, and 10-15% is performed at very high intensities, has been suggested as an optimal training intensity distribution for elite athletes who perform intense exercise events."  

For cyclists who race every weekend, I would suggest doing only one interval session per week in order to avoid over-training and consider the weekend races collectively as a second interval.  There are a number of bike racers who use racing to get themselves into shape.  This is tried and true, but I speculate that controlled intervals, bi-weekly as described in Table 2 (below) may be superior.   This may be because while racing will certainly produce stress that can trigger fitness, the efforts produced during a race may not be as completely exhausting as an ideal interval session or with the optimal amounts of intensity and work duration.  And because racing has little to no rest opportunities, most cyclists are likely to race more strategically-minded than HIT-minded. 


1.  Hickson RC, Bomze HA, Holloszy JO. Linear increase in aerobic power induced by a strenuous program of endurance exercise
2. Henritze J, Weltman A, Schurrer RL, et al. Effects of training at and above the lactate threshold on the lactate threshold and maximal oxygen uptake
3.  Simoneau JA, Lortie G, Boulay MR, et al. Human skeletal muscle fiber type alteration with high-intensity intermittent training
4.  Simoneau JA, Lortie G, Boulay MR, et al. Effects of two high-intensity intermittent training programs interspaced by detraining on human skeletal muscle and performance
5.  Green HJ, Fraser IG. Differential effects of exercise intensity on serum uric acid concentration
6.  Nevill ME, Boobis LH, Brooks S, et al.  Effect of training on muscle metabolism during treadmill sprinting
7.  Keith SP, Jacobs I, McLellan TM. Adaptations to training at the individual anaerobic threshold
8.  Linossier MT, Dennis C, Dormois D, et al.  Ergometric and metabolic adaptation to a 5-s sprint interval training
9.  Burke J, Thayer R, Belcamino M. Comparison of effects of two interval-training programmes on lactate and ventilatory thresholds
10.  Lindsay FH, Hawley JA, Myburgh KH, et al. Improved athletic performance in highly trained cyclists after interval training
11.  Tabata I, Mishimura K, Kouzaki M, et al.  Effects of moderate intensity endurance and high-intensity intermittent training on anaerobic capacity
12.  Westgarth-Taylor C, Hawley JA, Rickard S, et al.  Metabolic and performance adaptations to interval training in endurance trained cyclists
13.  MacDougall JD, Hicks AL, MacDonald JR, et al.  Muscle performance and enzymatic adaptations to sprint interval training
14.  Ray CA.  Sympathetic adaptations to one-legged training
15.  Green H, Tupling R, Roy B, et al.  Adaptations in skeletal muscle exercise metabolism to a sustained session of heavy intermittent exercise
16.  Rodas G, Ventura Jl, Dadefau JA, et al.  A short training programme for the rapid improvement of both aerobic and anaerobic metabolism
17.  Parra J, Cadefau JA, Rodas G, et al.  The distribution of rest periods affects performance and adaptations of energy metabolism induced by high-intensity training in human muscle
18.  Harmer AR, McKenna MJ, Sutton JR, et al.  Skeletal muscle metabolic and ionic adaptations during intense exercise following sprint training in humans
19.  Laursen PB, Shing CM, Peake JM, et al.  Interval training program optimization in highly trained endurance cyclists
20.  Laursen PB, Blanchard MA, Jenkins DG  Acute high-intensity interval training improves Tvent and peak power output in highly trained males
21.  Laursen PB, Shing CM, Peake JM, et al.  Influence of high-intensity interval training on adaptations in well-trained cyclists
22. Spencer MR, Gastin PB Energy system contribution during 200- to 1500- m running in highly trained athletes
23.  Laursen PB Training for intense exercise performance:  high-intensity or high-volume training? 




Sunday, February 5, 2012

The Ideal Cadence for Competitive Bicycling

Some time ago I wrote an article about Myths in Cycling regarding crank arm length and cited Dr. James Martin's powerpoint entitled, ""Myth and Science in Cycling:  Crank Length and Pedaling Technique"
As a result, I touched upon the topic of "ideal cadence", which has lead me to pursue this topic more thoroughly.

Sprinting:
Based on Dr. Martin's research and others, an ideal cadence for sprinting is around 120 rpm.  This isn't too surprising or controversial.  Sprinting is pretty straight forward.  Just pure power in a short burst of usually 5 to 15 seconds.  The side graphic shows this.  However, the myth that he busts isn't on cadence's effect on sprinting; it's on standard crank arm length's effect on sprinting (basically there's very little difference between standard crank sizes on performance).


Gear Choice:
 Another interesting point about gear choice (which affects cadence) for sprinting  is that for longer sprints such as 30 seconds versus 10 to 15 seconds, a bigger gear choice is desirable.  Research shows that the reason for this is  that muscle fatigue occurs more due to the total number of muscle contractions than by the duration of the contractions during extended maximal efforts.    Basically it's more ideal to grind a bigger gear on a long sprint than to be "spun out", especially when another gear choice is available.

Below is a gear chart of speeds (via Sheldon Brown's Gear Calculator) produced using a 52 and a 53 front chain ring and a 10 speed cassette ranging from 11 teeth to 21 teeth.  On a flat course most riders would sprint well in a 53/52 x 14 or 13.  Professional riders may sprint in a 12 rear cog, and Kevin Sireau may use an 11 to sprint with. 
Cadence:
The information about crank-arm length from Dr. Martin is pretty clear. Standard sizes are nearly equally efficient, but the topic of ideal cadence came up and it generated a bit of controversy among my peers.  I posted a data graphic that illustrated that 60 rpm cadence (pedal revolutions per minute) was more efficient than 100 rpm.  The myth that this busted was that higher cadences are better for performance than lower.

This flies in the face of the dogma that higher cadence is better, something that I have been told since I began cycling competitively.   I was concerned about this contradictory information to the point that I  contacted Dr. Martin via email for clarification.  He confirmed the facts as such:  "The effect of pedaling rate on metabolic cost is pretty well established. Heart rate generally tracks well with met cost but its not the same thing. Also, there is individual variability in responses so you may be a bit different than the mean."  (Cool, Dr. Martin!)

But hold the door, this isn't entirely settled.  Several of my peers have challenged this idea.  They point to examples of Lance Armstrong time-trialing at 100 plus cadence and beating the competition and the fact that The Cycling Hour World Record has been set with cadences above 100.  Add to that the fact that most professional and elite athletes will generally ride at higher cadences (90-105) while racing as well.  This naturally leads one to think that these riders must be riding at the ideal cadence and that  lower cadences such as 60 is just wrong. 

So I set out to resolve this, and here's what I found:  Both are true.  Basically.  I looked at dozens of research papers (see sources below) on the effects of cadence and efficiency and found over and over that Dr. Martin was right, but with a caveat.  Lower cadences are more efficient for the vast majority of normal riders who do not have huge aerobic capacities and can not sustain large power outputs. 

Two possible explanations for why efficiency and higher cadence numbers go up as power output goes up for elite and professional riders.  One is heat generation within the muscle fibers and the other is muscle composition.  Cyclists across the board are generally only about 24% efficient when pedaling.  Efficiency is a measure of work performed for energy used.  Or for cyclists we'll say, less oxygen used, less fuel burned, means more efficient.

Net mechanical efficiency for muscle movements is generally low for all cyclists due to the loss of free energy as heat.   As power output increases, so does the temperature within the muscle.  This temperature factor may have an effect on efficiency with the speed at which a muscle contracts (or cadence).  Additionally, it matters which type of muscle fiber that is contracting.  Fast twitch and slow twitch fibers have different contractile properties in terms of efficiency and optimal speed of contracting.  As work load (or power output) goes up, more and more muscles fibers are recruited to do work.  As a result of either-and-or heat build-up in the muscle fibers and the type of fiber being recruited, both efficiency and cadence go up as power output increases.  

Take a look at the graphic below.  It's a great illustration of how elite riders become more efficient with higher cadences at greater levels of power output.  

This data comes from Øivind Foss and Jostein Hallén's article, "The most economical cadence increases with increasing workload"  It is extremely important to note that the six rider's in this study are elite cyclists with a VO2 max of an average of 69 ml/kg/min which puts them far outside of  the power output of most cyclists.  These elite riders are capable of sustaining 350 watts for an hour (time trialing around 29mph) whereas a cat 5 or untrained rider can only sustain from 130 to 200 watts for an hour (time trialing around 20-23 mph).  Big difference.

Cadence During Races
Now lets discuss cadence in practical terms regarding racing.  Cadence should largely be ignored during criterium racing.  No more thought should be applied to cadence in criterium racing than to what your respiratory rate is (that is zero).  The focus should be staying close; tight in the draft of the cyclist in front of you at all cost.  The reason for this is that the efficiency of drafting can be upwards to 36% energy savings whereas ideal cadence may only be single digit percent savings of energy.   With that said, it is also worth mentioning that the nature of criterium racing is very much "gas on- gas off" (full power and zero pedaling), high cadences are easier for accelerating and a bit easier on your connective tissues in your knee and leg (especially with long riding without rest or variation).  Again, for criterium racing, focus on drafting and let cadence take care of itself. 

Ideal cadence is more important for time trialing, where every second counts, because it's a race against the clock.  In the case of the elite cyclists above, clearly an 80 cadence is the most efficient to ride at (coincidentally 350 watts happens to be their average lactate threshold).  Time trialing at a lesser efficient cadence of 60 or 100 would cost approximately 9 watts of power (my estimate) and would result in about a 30 second time difference over a 40 kilometer time trial.  (my calculations come from  the American College of Sports Medicine formula  and the bike calculator).  So an ideal cadence is important.

But here's the problem.  It is not possible to maintain exactly 80 rpm while time trialing at a maximal effort in the real world for several reasons.   Firstly, there is a fairly big change in the power  requirement between gear changes at a fixed cadence.    See the gear chart below to see what I mean.  There's about a 2.1 to a 2.8 mph difference between gear changes, which translates to a difference of nearly 100 watts of power in the upper gear sizes. 
Additionally there are several other factors beyond power that affect you on the road, such as wind changes and road grade changes.  As a result, cadence (and often gearing on hilly courses) must constantly be adjusted for these factors.

The elite riders studied in Øivind Foss and Jostein Hallén's article, "The most economical cadence increases with increasing workload"  (first chart) have an average  VO2 max of 69 and a weight of 78 kg (or 171 lbs) and as a result we would expect them to ride a 40k time-trial at about 28.65 mph (full aero equipment, flat terrain, zero wind, 100ft elevation).  There's no data point on the above gear chart for an 80 cadence gear combination that gives us this exact speed.  However the 90 cadence gear chart (above left) does put our riders very close to the speed of 28.65 with the gear combination of 53 x 13.

Of course, we know from our earlier discussion that 80 rpm cadence is around 9 watts and 30 seconds faster for our riders than a 90 cadence.  So what can be done?  The best that can be done (for our elite riders) riding with a 53 tooth chain ring is to select a 12 rear cog and pedal at a slightly elevated cadence of 83 rpm.  The 9 watts and 30 seconds advantage will be reduced proportionally (est. 6w, 20s).  This is far easier done in a lab than in the real world!

A Real-Life Example:
A good example of real world riding is illustrated in the SRM data from my 2010 State Time Trial, which was raced on a flat course.   I did not focus on cadence, but instead focused on my effort, shooting for an average of 345 watts.    Because of the changing wind conditions, my speed and cadence would go up and down, but I maintained a fairly consistent power output.   Looking at this graph with my current knowledge, I may have had a bit of a benefit from a slightly higher cadence in the later half.

[It is worth knowing that it is generally best to be the first rider to start on an out-and-back course in the morning, and the last rider to go in a time-trial in the late afternoon, because the wind tends to pick up in the morning and tends to die down in the evening.   Time trials with large numbers of competitors with one minute intervals in between, can lead to hours of differences in start times that will likely have different environmental conditions.    Zero wind is the most ideal for performance for out-and-back time trials.]

Below:  The purple is speed, blue is cadence and green lines are wattage.  The drop down spike in the center is the turn around point.  
click to enlarge view
Efficiency can be rather complicated to determine on the fly.  Happily, heart rate tracks very well with efficiency (see graphic below) for work done.  Therefore, one should try to maximize power and speed output with the lowest heart rate possible so as to achieve the best efficiency and greatest production (fastest time trial)
The above graphic shows the relationship between cadence,  heart-rate and power and comes from the same study of the elite riders as mentioned above

I looked at several more studies regarding ideal cadences for cycling and they are worth mentioning and looking at and include:

Effects of Altering Pedal Cadence on Cycling Time-Trial Performance
Found that low cadence 83 was both faster and more economical than preferred cadence of 92 

Cadence and performance in elite cyclists
Found that elite cyclists performed best and was more efficient time trialing at 80 cadence


Preferred pedaling cadence in professional cycling
Found that professional riders spontaneously adopt higher cadences (around 90) during both time trialing and group riding, but tend to adopt a more economical pedaling rate of approximately 70 rpm during hill climbs.

 Effect of cadence, cycling experience, and aerobic power on delta efficiency during cycling
Found that little difference exists between trained and untrained cyclists concerning efficiency (usually around 24%) regardless of cycling experience or fitness level


In Professional Road Cyclist, Low Pedaling Cadence Are Less Efficient
Found that professional road cyclist riding at power outputs greater than 360 and 420 watts are more efficient at 100 rpm than 60 and 80 rpm


Cycling efficiency and pedalling frequency in road cyclists
Efficiency increases in scale with pedaling rate as workload increase.


Cadence and performance in elite cyclist
This study demonstrated that elite cyclists perform best at their most efficient cadence which was 80 rpm,  despite the maximal energy turnover rate being larger at a higher cadence.

In short, you should  choose a cadence that mirrors your power output; unless you’re an elite rider, it’s unlikely you’ll benefit from using cadences exceeding around 80rpm.  However, world-class athletes can push into 100 rpm range for the most efficient cadence that will produce the greatest performance.  

To see where you stack up in the field of competitive cyclists, see my post:  Comparative Measurement of Maximal Outputs for Cyclists.